physiological and endocrinologicalresponses

Acta vet. scand. 2000, 41,173-184.

Physiological and Endocrinological ResponsesDuring Prolonged Exercise in Hatchery-RearedRainbow Trout (Oncorhynchus mykiss)

ByM. E. Nielsen', L. Boesgaard' , R.M. Sweeting', B. A. McKeown2 and P. Rosenkilde'

'Department ofVeterinary Microbiology, Section ofFish Diseases, The Royal Veterinary and Agricultural University, Frederiksberg, Denmark, 2Department of Biological Sciences, Simon Fraser University, Burnaby, British Columbia, Canada, 3Zoophysiological Laboratory, August Krogh Institute , University ofCopenhagen, Denmark.

Nielsen ME, Boesgaard L, Sweeting RM, McKeown BA, Rosenkilde P: Physiological and endocrinological responses during prolonged exercise in hatchery-rearedrainbow trout (oncorhynchus mykiss). Acta vet. scand. 2000, 41, 173-184. - Rainbowtrout (Oncorhynchus mykiss) were subjected to vigorous exercise (1.5 body length S-I),low exercise (0.5 body length S·I) or still-water (0.0 body length S·I). Hematocrit, glucose, growth hormone (GH), cortisol and triiodo-L-thyronine (T3) were monitored at thestart of exercise, after 24 h, and after 2, 4, 8, 16 and 32 days of continuous swimming.Morphological indices and food intake were also monitored. At the end of the experiment, trout subjected to low exercise had gained significantly (p<0.05) more weight thanboth the control (still-water) and vigorously exercised fish. This low exercised group offish also consumed more food than the 2 other groups . Hematocrit increased significantly in both exercised groups at the onset of swimming but returned to pre-exerciselevels within 8 hrs. Plasma glucose appeared to be generally unaffected by exercise.Likewise, plasma concentrations of both GH and T3 were not influenced by exercise.Plasma cortisol levels increased in an exercise dependent fashion at the onset of swimming, but returned to pre-swimming levels within 24 h and there was no apparent effectof sustained swimming . The results suggest: (i) the onset of exercise elicits transientchanges in plasma components, (ii) the observed weight gain in low exercising salmonids occur without increases in circulating levels ofGH or T3"

Cortisol; Growth hormone; T3; Hematocrit; Fish growth; Exercise; Rainbow trout.

IntroductionDuring the last decades, several studies havedemonstrated that salmonids continuously exercised at modest speeds (1.0 - 2.0 body lengthssec"; BLS) may show increased growth compared to non-exercised controls (Walker &Emerson 1978,Houlihan & Laurent 1987, Totland et al. 1987, Christiansen & Jobling 1990,Davidson 1997). In addition to the generallyimproved fitness (East & Magnan 1987), the increased growthobserved in these and other

studies has been attributed to increases in bothfood consumption (Leon 1986, Tot/and et al.1987) and food conversion efficiency (Kuipers1982, Leon 1986, Davidson 1997). However,while several experiments have dealt with thesegeneral factors, few have investigated thechanges in hormonal and other regulatory factors during long-term exercise.Increase in appetite or food consumption andfood conversion efficiency are 2 functions

Acta vet. scand. vol. 41 no. 2, 2000

174 M. E. Nielsen et al.

which, in teleosts as well as in other vertebrates,are known to be modulated by the growth hormone (GH) (Weatherley & Gill, 1989, Bjornsson 1997) in addition to its effect on growth andprotein synthesis (McKeown et al. 1975, Weatherley & Gill 1982, Gill et al. 1985, McLean etal. 1990, Foster et al. 1991). Furthermore, during sustained swimming, rainbow trout (Oncorhynchus mykiss) preferentially mobilize lipidreserves and conserve body protein (Beamish1978), and several studies have demonstratedthe important lipolytic role played by thegrowth hormone in salmonids (McKeown et al.1975, Sheridan 1986, 1994, Bjornsson 1997).Finally Barrett & McKeown (1988) and Kakizawa et al. (1995) have shown that plasma concentrations of the growth hormone were highlyelevated in steelhead salmon (Oncorhynchusmykiss) following 24 h of exercise at highspeeds (1.5 BLS). However, these studies measured plasma hormone concentrations post-exercise and little information exists on growth hormone levels during either short term or longterm exercise in fish.Although growth promotion has most frequently been associated with GH, thyroid hormone is also involved in growth promotion(Higgs et al. 1977). Furthermore, 3,5,3'-triiodo-L-thyronine (TJ), the more potent form ofthe thyroid hormone (Eales 1985), has beenshown to stimulate transcription ofgrowth hormone mRNA (Moav & McKeown 1992) andthus, indirectly, stimulate growth. GH has alsobeen implicated in the regulation ofT, production (MacLatchy et al. 1992). Thus thyroid andgrowth hormone may be interacting in exercised fish. However, there is limited information in the literature on links between exercise,growth and the thyroid hormone .In contrast to the anabolic actions of bothgrowth hormone and T), the actions of cortisolare generally considered to be catabolic andgrowth suppressive (Barton et al. 1987,Picker-


ing 1990). The imposition ofan exercise regimeon fish might be considered a stress challenge(Boesgaard et al. 1993) the degree dependingupon the velocity ofthe water current the fish isforced to swim against (Niels en et al. 1994).Furthermore, the catabolic actions of cortisolmay provide sources ofenergy for the increasedmetabolism associated with swimming againsteven a moderate current. A few studies have examined the effects of exercise on subsequentstress challenges , for instance osmoregulation(Postlethwait & McDonald 1995, Mommsen1999), but little is known about the possible alterations in plasma cortisol levels.The aim of this study was to investigate possible correlations between improved growthand involved hormonal factors (GH, T3 andcortisol) in rainbow trout subjected to vigorousexercise (1.5 BLS), low exercise (0.5 BLS), orstill-water (0.0 BLS) for a period of 32 days.The level and length of the exercise periodewere chosen to prevent induction of anaerobicwhite muscle work (Johnston & Moon 1980).Whereas most of the previous work on exercisein fish has focused on post-exercise effects, thisstudy was designed to obtain better insight intometabolic and endocrine parameters during lowand vigorous exercise. Therefore , food intakeand growth (weight, length and condition factor) were correlated with levels of GH, thyroidhormone and cortisol plus hematocrite andblood glucose . The food intake of the fish wascarefully measured during the study to determine possible changes in appetite .

Materials and methodsFishJuvenile I+ rainbow trout (n = 540); meanweight 94.5 ± l.3g; mean length 20.4 ± 0.1 emfork length) were obtained from the West CreekHatchery, Langley, British Columbia, Canada,from a strain selected for 8 generations for fastgrowth in a low water current environment (A.

Responses during prolonged exercise in rainbow trout 175

Ludwig, pers. comm). The fish were distributedequally (n = 160) in three 2000 I fibreglass ovalraceways and acclimated for 2 weeks under natural photoperiod and temperature (10-11Qq .Each racewaywas supplied with a constant flowof oxygenated and de-chlorinated tap water(300-L h· I ) . The raceways had 2 straightfenced-in sections (each with a volume of 600I) in which trout (n = 80) were kept (McKinnon& Farrell 1992). During the two-week acclimation period, the fish were hand fed I% of bodyweight per day with commercial trout pellets(Nelsons Sterling Silver Cup).On day 0 of the study,water flowwas created bya 24-volt electric outboard motor in 2 of the 3raceways. One raceway had a water current setat approximately 0.1 m S·1 (corresponding to0.5 BLS) while the other raceway had a watercurrent ofapproximately 0.3 m S·I (corresponding to 1.5 BLS), subsequently referred to as thelow and the vigorously exercised groups, respectively. The fish referred to as the still-wateror control group were maintained in standingwater (0.0 BLS). During the 32 days of exercise, the fish ofeach raceway were carefully fedto satiation twice daily. In this way it was possible to estimate the amount of food given toeach group and thereby get an overall impression of appetite .

Raceway designIn designing the chambers and raceways, therewas a conscious attempt to minimise turbidityor other variations in water-flow (McKinnon &Farrell 1992). It should be noted that while thestill-water fish were not forced to swim againsta current, these fish exhibited normal randomswimming tendencies and thus cannot be designated as non-active.

SamplingSampling occurred immediately before the onset ofexercise (time = 0), at 1,2,4, 8 and 24 hrs,

and after 2, 4, 8, 16 and 32 days of exercise. Tominimise any effects of diurnal changes onplasma parameters, sampling always started at9 a.m. (except over the initial 24 h). At eachsampling time, 8 fishwere rapidly netted and lethally anaesthetised with neutralised MS-222(Syndel Labs., Be., Canada). All fish were unconscious within 10-15 sec. Blood was collected from the caudal vein in heparinized syringes, dispensed into 1.5 ml microcentrifugetubes and immediately placed on ice. Completesampling procedure (netting, weights and bloodcollection of all fish) was performed within 810min. Blood samples were centrifuged (5 minx 16000 g), hematocrits recorded and plasmacomponents stored at -20QC until further analysis. Fork length, liver weights and mesentericfat content were obtained following completionof sampling (carcasses stored on ice). Extremecare was taken to minimise stress during eachof the sampling periods. Furthermore , fish fromeach consecutive sampling were taken from thealternate chamber of the appropriate raceway,thus allowing a period of recovery for the remaining fish and minimising stress from theprevious sampling period.

AnalysisBody wet weights and fork lengths were obtained for all fish at each sampling period.Mean liver weight and weight of mesenteric fatwas determined at 0, 2, 4, 6, 16 and 32 days.Plasma growth hormone levels were determined in triplicates by radio-immuno-assay(RIA) for salmon GH essentially as describedby Sweeting et al. (1985) but with the followingdifferences: (i) both the GH standards and theradiolabelled [125I]_GH were stored in acetatebuffer (O.2M with 0.5% BSA; pH 5.0), (ii) thefinal precipitating buffer contained 4.0% PEG(polyethylene glycol, Sigma, USA). Plasma3.5.3' -triiodo-L-thyronine was determined induplicates by a commercial RIA (Gamma-



Table I . Morphological data for rainbow trout exercised at 0.0 (Control), 0.5 (Low exercise) or 1.5 (Vigorousexercise) body lengths sec'] for 32 days . Each value represents means (n = 8) ± SEM. (*) Represent a significantdifference from Day O. Values with different letters are significantly different from each other.

Day0 Day32

Group Weight (gms) Length (em) Weight (gms) Length(em)

Controls 98.46 (4.18) 21.29 (0.17) 119.2' (14.14) 21.88' (0.57)Low Exercise 96.07 (3.28) 20.76 (0.27) I54.29'b (10.51) 22.78' (0.68)Vigorous Exercise 96.83 (4.34) 21.63 (0.25) 111.29"(12.05) 21.48 ' (0.39)

CoatTM [1251] T3 RIA Kit, INCSTARCorporation, USA) The limit of detection of this RIAwas 0.5 ngml' . Plasma cortisol values were determined in triplicates using a commercial ELISA kit (Diagnostic Systemes Laboratories,USA) with a detection limit at 2 ng ml' . Duplicates of plasma glucose titers were analysedcolorimetrically (#16UV, Sigma, USA).

Statistical analysisData were analysed using two-way analysis ofvariance (ANOVA, Sheffe F-test) with treatment and time as factors. In each group postexercise values (32 days)were compared to preexercise values (0 days) using Student's t-test.Significant differences were accepted at the

probability level of 0.05. All data are given asarithmetic means ± S.E.M.

ResultsNo mortality was observed in any of the exercise groups throughout the study. Furthermore,no fish were at any time obviously fatigued bythe continuous enforced exercise. Thus, therewere no observations of fish "resting" againstthe back of the holding chamber.All 3 groups showed a significant increase inlength from 0 to 32 days (Table 1), but therewere no significant differences between thegroups at 32 days. Both exercise groups showedsignificant increases in weight from 0 to 32days, but the low exercised fish had signifi-

1.5

1.4

1.3

I 1.2

<Z 1.1

-ci6 1.0u

0.9

0.8

3.0

2.5

g 2.0

1.5

1.0::;

0.5

N'

0.0 -',------,- --,-- --.-------,--

o 16Days

24 32 16

Days

24 32

Figure I . Condition factor (w/l ') (mean ± SEM) .(0) Control group (n = 6-8); (e) Low exercised trout(n = 8); (\7) Vigorously exercised trout (n =8).

Acta vet. seand. vol.41no.2, 2000

Figure 2 . Liver weight in gram (mean ± SEM) . (0)Control group (n = 8); (e) Low exercised trout (n =

8); (\7) Vigorously exercised trout (n = 8).


1.17

L 1.31

- .:l-' c",;:; 1.01

1.04j[

cl 0.88

.'

I //f' . /.:7 ". / . . . . .

:' /- f- - - -

y" '!

.'//,t

/

k l?'ICj" , .r ' / j.... . .

0.90 ..L,.- - - -,- - - -.- - - .,-- --, O..L,.- - - --,- - - --,- - ---,--- -..16

Days

24 32 o 16

Days

24 32

Figure 3. Hepatosomatic index (liverweight in percent ofbody weight). (0) Control group (n = 8); (e)Low exercised trout (n = 8); (\7) Vigorously exercised trout (n = 8). (*) Different from vigorouslyexercised trout and control group. (#) Represent asignificant difference from Day O.

Figure 4 . Mesenteric fat weight in gram (mean ±SEM). (0) Control group (n = 8); (e) Low exercisedtrout (n = 8); (\7) Vigorously exercised trout (n = 8).(*) Different from vigorously exercised trout andcontrol group. (d) Different from low exercised troutand control group . (#) Represents a significant difference from Day O.

3.0 , ,Table 2 . Average weight gain, food consumption

2.5 /// t and utilisation of rainbow trout for rainbow trout ex-ercised at 0.0 (Control) , 0.5 (Low exercise) or J.5

- 11 2.0 (Vigorous exercise) body lengths sec') for 32 days.>-"' 1.5"8 Weight Foode:; E

Group Gain EatenUtilisation

j 1.0 (IOOO*wt/food),, ::;; ····t· ······· ··· (gms) (gms)!;,

0.5 il

Controls 20.7 36 0.5780.0 Low Exercise 58.2 40 1.437

0 16 24 32 Vigorous Exercise 14.5 30 0.480Days

Figure 5. Mesenteric fat index (mesenteric fat index in percent of body weight). (0) Control group(n=8); (e ) Low exercised trout (n = 8); (\7) Vigorously exercised trout (n = 8). (*) Different from vigorously exercised trout and control group . (d) Different from low exercised trout and control group. (#)Represent a significant difference from Day O.

cantly higher mean weights at the end of thestudy compared to either the still-water groupor the vigorously exercised trout. Besides theeffects seen directly from increments in lengthor weight, the condition factor suggestschanges in the distribution of the gained bodymass (Fig I) .

Liver weight (Fig. 2) and the hepatosomatic index (HSI) (Fig. 3) increased during the study.Following 32 days of prolonged exercise thefish of the low exercise group had significantlyhigher liver weight and HSI compared to theother two groups. Furthermore, the liver weightand the HSI in the low exercised fish after 32days were significantly higher than pre-exercised values in this group.As seen with the other parameters , fish experiencing 32 days of exercise at low level exhibit asignificant increase in both total weight ofmesenteric fat as well as in percentage mesenteric

Acta vet. scand . vol. 41 no. 2. 2000


20

O ...J..,- - - ---,- - - ---,- - - ---,- - - ---,-

0.030

f0.025

s1 0.020

'"0.015

]§ 0.010

0.005

0.000

16

Days

24 32

16

E! 12

6E 8

0:

o 16

Days24 32

Figure 6 . Food consumption (gram food consumed/gram fish/day) fitted curve . (0) Contro lgroup; (e) Low exercised trout; (\7) Vigorously exercised trout.

Figure 7. Plasma GH (mean ± SEM). (0) Controlgroup (n=8); (e) Low exercised trout (n=8); (\7)Vigorously exercised trout (n = 8). (*) Different fromlow exercised trout and control group . (#) Representa significant difference from Day O.

10

/t- ·fr! /" .i· " /" .

"' ." y./ . . .

225

200

E 175

! 150

125'e0u 100

750:

50

25

B

-=- 150

'e 1008E 50

E 0c,

.__ ·

o 12 16 20 24Hours

16

Days

24 32 o 16Days

24 32

Figure 8 . Plasma T3 (mean ± SEM) . (0) Controlgroup (n = 8); (e) Low exercised trout (n = 8); (\7)Vigorously exercised trout (n = 8).

fat to body weight (Fig. 4 and 5). Mean valuesof both these parameters stayed level for thecontrol fish, but those fish forced to swim at 1.5BLS showed a large and significant decrease inthis parameter at 16 days of exercise which wasapparently restored by 32 days.Appetite and food conversion was used as anindirect expression of growth response to exercise. The moderately exercised fish consumedmore food than either the controls (113%) or


Figure 9 . Plasma cortisol (mean ± SEM). (A) Initial reactions . (B) Prolonged reactions (0) Controlgroup (n = 8); (e) Low exercised trout (n = 8); (\7)Vigorously exercised trout (n = 8). (*) Different fromcontrol group .

the high exercise group (134%) (Table 2). Furthermore, food conversion efficiency (definedas 1000 x weight gain in grams/food consumedin grams) over the 32 days was some threefoldgreater for the fish exercised at 0.5 BLS. Thefish forced to swim at 1.5 BLS, on the otherhand, showed less appetite and lower feed conversion than did the controls . Calculation offood consumption revealed that fish swum at1.5 BLS initially showed a decrease in appetite


DiscussionWootton (1990) defines appetite as "the quantity of food consumed before the fish ceases toeat voluntarily". The results from this study indicate that, for this stock of rainbow trout, lowexercise increases both appetite (=-13%) and thegrowth/food intake ratio or feed conversion(.,threefold) over that of still-water controls orfish kept at 1.5 BLS. Thus leading to a significantly greater increase in weight gain than observed in the other 2 groups . Furthermore, thevigorously exercised fish exhibited both lowerappetites and lower feed conversion than did thelow exercised fish or the control fish. The lessfood taken by the vigorously exercised groupsuggests that the higher bioenergetic demandsfaced by these fish are, in some fashion, decreasing either appetite or feeding behaviour. Itis possible that the additional effort of obtainingthe food while maintaining position (or returning to position) somehow reduces the appetitein some way. These results are in accordancewith earlier works which has reported that juvenile Atlantic salmon forced to exercise exhibited greater growth, food intake and food utilisation than did still-water controls (Jorgensen& Jobling 1993, Jorgensen et al. 1996). In contrast fish swum at 2.0 BLS had lower values forthese parameters than both controls and theother exercised groups. Besner (1980) alsonoted that increased water velocities did increase growth rates in juvenile coho salmon,and that further increase in water velocity (2.0BLS) resulted in decreased growth rates. Therainbow trout used in this study were from astrain which had not been exposed to streamingwater for some generations , and this may explain why lower growth was observed at relatively lower swimming speeds (1.5 vs. 2.0BLS). On the other hand, there are probably realdifferences in the physical capabilities betweenvarious salmonid species, which would be reflected in their ability to enhance growth under

Acta vet. scand. vol. 41 no. 2. 2000

different swimming regimes (White & Li 1985) .Further, both species and size differences mayhave contributed to differences in growth response to exercise. (Beamish 1978, White & Li1985). The increase in liver weights during thestudy corresponds well with the increase inbody weight.The low exercised fish exhibited a significantincrease in percentage ofmesenteric fat after 32days ofswimming, both as total weight and as apercentage of body weight. Both the controlsand the high-exercised fish also showed someincrease in total mesenteric fat over the sameperiod, but not when expressed as a percentageof body weight. Increase in lipid deposition inmuscle during prolonged exercise has been observed in previous studies with brown trout(Davison & Goldspink 1977) and brook trout(East & Magnan 1987). To our knowledge, thisis the first report of increased lipid deposition inmesenteric tissues during exercise of fish.These data suggest that ration was not limitingfor any of the 3 groups and that lipid depositioncontinued even in the fish undergoing high exercise. The increase in mesenteric fat content inthe low exercised fish probably contributed tothe observed increase in condition factor. Increase in condition factor has previously beenfound in fish undergoing prolonged exercise.For example , Jorgensen & Jobling (1993) notedthat while the characteristic smoltification-associated decline in condition factor was seen inthe control fish, all groups of Atlantic salmonundergoing forced exercise exhibited an increase in condition factor. In contrast, Kuipers(1982) reported that sustained exercise of postsmolt Atlantic salmon resulted in decrease incondition factor. The effects ofexercise on condition factor may well depend upon species andespecially life stage. It is apparent that much ofthe increase in mesenteric fat and condition factor occurred during the last sampling period(i.e. days 16 to 32). An extension of the swim-


55 B 55 A

5050 I 45

40 ;/s 45 :l! ..' ?' C

35

40 8 12 16 20 24

:l! Hours

35 - - - - -

30

0 16 24 32Days

Figure 10 . Hematocrit (mean ± SEM) . (A) Initialreactions. (B) Prolonged reactions (0) Contro l group(n = 8); (e) Low exercised trout (n = 8); ('7) Vigorously exercised trout (n = 8). (*) Different from control group .

which at the end of the study increased dramatically to a level similar to the low exercised fish(Fig. 6).There was a slight increase in the levels ofplasma growth hormone during the study (Fig.7). The mean concentrations in the 0.5 BLS fishafter 32 days ofcontinuous swimming were significantly elevated over pre-exercised values aswell as at all other times this parameter was analysed for this group. The plasma GH level wasalso significantly higher in the 1.5 BLS fish at32 days than in the pre-exercised fish. However,none of the plasma GH values in the threegroups were significantly different from eachother at this time or any other time dur ing thestudy, with the exception ofa single peak in GHin the vigorously exercised fish on day 8. (Note :Plasma GH was not determined over the initial24 h of the study due to previous commitmentsof the plasma).The plasma levels ofT, in the exercised groupsand the contro l group showed a non-significantincrease during the study (Fig . 8). The non-ex ercised group had a more rapid elevation inplasma T3 than both exercised groups. Betweenthe exercised groups the low exercised group

had the fastest elevation in T3 during the experiment.The greatest impac t of exercise was seen forcortisol (Fig. 9), which rose in all 3 groups following the start of the experiment. In the control fish, increase in plasma cortisol was veryslight and short (presumably due to samplingstress), with the only significant differen t valuesseen at I h (higher than both 0 and 8 h). In thegroup offish forced to swim 1.5 BLS, there wasan extremely high and significant increase inplasma cortisol within I h (186 .25 ng ml! vs.23.13 ng ml' at time 0). The plasma cortisollevels continued to be significantly elevated andhigher than both control and low exercise fish at2 and 4 h post -onset. After 8 h the plasma cortiso l concentration had returned to normal postexperimental values . Plasma cortisol levels alsorose in the low exercise fish after the onset ofexercise , but only the concentrations at the 4and 8 h sampling periods were significantlyhigher than in the contro ls. Plasma cortisol levels had returned to pre-exercise levels within 24h ofthe onset ofexercise , and there were no apparent effects of exercise on plasma cortiso llevels during the rest of the study. There was alarge but not read ily explained significant peakin plasma cortisol levels in the contro l group offish at the 8 day mark .Hematocrits increased significantly within Ihatthe onset of exercise in both the low and vigorously exercised groups (Fig. 10). Furthermore,the increase in hematocrit in the vigorously exercised fish was significantly higher than the increase seen in the low exercised fish. By 8 h ofexercise, the hematocrit values in all 3 groupshad returned to pre-exercise values , where theyremained till the end of the study. There wereno changes in plasma glucose concentrationsassociated with either exercise or samp lingthroughout the 32 days ofexercise, although thevigorously exercised group consistently exhib ited higher mean values (data not shown) .

Acta vet. scand . vol. 41 no. 2, 2000


ming regime over another sampling period mayhave shed further light on these results.The results from the present study show thatneither plasma GH or plasma T3 were elevatedin rainbow trout during the 32 days of continuous exercise at either low or vigorous rates.Forsman & Virtanen (1989) found that insmolting Atlantic salmon, exercise (2.5 BLS)led to a transient increase in plasma thyroxine,which had returned to normal levels within 12 hof onset. Based on similar changes in plasmacortisol , the authors concluded that at this speedthe fish were primarily swimming anaerobicallyand, particularly for the precocious males,under some stress (Forsman & Virtanen 1989).In a companion study utilising the same system,same strain of rainbow trout and similar watervelocities as in the current one, a non-significant increase in circulating T3 levels over 24 hof exercise were noted (Nielsen et al. 1994).Consequently there seems to be no obvious relationship between the described increase ingrowth observed in the exercising salmonidsand the plasma levels of these hormonal factors. A possible explanation for the observedincrease in plasma T3 in the low exercisedgroup and the still-water group could be thehigher food consumption in these groups , as aclose connection between increased food intakeand elevated plasma T3 levels has been reported(Eales 1988, Sweeting & Eales 1992).The studies of Barrett &McKeown (1988) andKakizawa et al. (1995) demonstrated increasein plasma growth hormone concentrationwithin 5-10 min post-exercise if fish had beenexercised for at least 4h (Barrett & McKeown1988). The exercise-associated increase inplasma GH was not seen during exercise, butrather after exercise had ceased . Nielsen et al.(1994) found a single transient peak in plasmaGH within 2 h ofonset ofvigorous exercise (1.5BLS), whereas no such peak was observed inthe low exercised (0.5 BLS) fish. It should be

noted that the post-exercise increases in plasmaGH demonstrated by Barrett & McKeown(1988) were observed within 10min ofexercisestoppage and had returned to normal levelswithin 2 h. The prolonged effects of exercise onthe plasma levels of growth hormone seem tobe a slight elevation during the experiment. Altogether, low exercise appears to form a favorable condition for growth in accordance withnumerous experiments, as reviewed by Bjornsson (1997).Plasma cortisol levels also exhibited a transientincrease at the onset of exercise, the magnitudeofwhich appeared to be dependent on the speedof swimming. The levels of circulating cortisolin both of the exercised groups returned to basal values within a very short time, and did notappear to be further affected by continued exercise. The lack of change in plasma glucose maysupport the contention that plasma cortisol titers are not changed by exercise per se. Transient increases in hematocrit following physicalstress have been reported previously (Soivio &Oikari 1976, Barton et al. 1987), lending further support to the hypothesis that it was the onset of exercise that stimulated the cortisol response rather than the exercise per se.

ConclusionIn conclusion, the results of this study demonstrate that low exercise (0.5 BLS) over a prolonged period can induce increased growth,whereas exercise at higher levels may inhibitgrowth, possibly via increased bioenergetic demands (Kiessling et al. 1994). The increase ingrowth were achieved by an increase in both appetite and food conversion efficiancy. The results suggest that exercise-induced weight gaincan occur in the absence of an obvious increasein circulating levels of growth hormone.Plasma T3 levels, on the other hand, seemed tobe more influenced by food intake than watercurrent. Plasma cortisol did not appear to be af-



fected by exercise per se, except for the onset ofexercise at which time both cortisol and growthhormone levels showed transient increase .

AcknowledgementsNatural Sciences and Engineering Research Councilof Canada grant A9434 awarded to Brian A.McKeown supported this study.

ReferencesBarrett BA, McKeown BA: Sustained exercise in

creases plasma growth hormone concentrationsin two anadromous salmonids. Can. J. Fish.Aquat. Sci. 1988,45,747-749.

Barton BA, Schreck CB, Barton LD: Effect ofchroniccortisol administration and daily acute stress ongrowth, physiological conditions, and stress responses in juvenile rainbow. Dis. Aquat. Org.1987,2,173-185.

Beam ish F.WH: Swimming capacity. In: Fish Physiology, vo!. VII. W.S. Hoar and DJ. Randall, eds.,Academic Press, New York . 1978.

Besner M: Endurance training : an affordable rearingstrategy to increase food conversion efficiency,stamina, growth and survival of coho salmonsmolts (Oncorhynchus kisutch). Ph.D dissertation, University of Washington, Seattle. 1980.

Bjornsson BT: The biology ofgrowth hormone : fromdaylight to dominance . Fish Physio!. Biochem.1997,17,9-24.

Boesgaard L, Nielsen ME and Rosenkilde P: Moderate exercise decreases plasma cortisol levels inAtlantic salmon (Salmo salar). CompoBiochem.Physio!. 1993, 106A, 641-643.

Christiansen JS, Jobling M: The behaviour and therelationship between food intake and growth ofjuvenile Arctic charr, Salvelinus alpinus L., subjected to sustained exercise. Can. J. Zoo!' 1990,68,2185-2191.

Davison W The effect of exercise training on teleostfish, a review of recent literature. Compo Biochern. Physio!. 1977, 1I7A, 67-75 .

Davison W. Goldspink G: The effect of prolonged exercise on the lateral musculature of the browntrout (Salmo trutta). J. Exp. Bio!. 1977, 70, 1-12.

Eales JG : The peripheral metabolism of thyroid hormones and regulation of thyroidal status in poikilotherms . C. J. Zoo!. 1985, 63, 1217-1231.

Eales JG : The influence ofnutritional state of thyroidfunction in various vertebrates . Am. J. Zoo!.1988,28,351-362.

Acta vet. scand . vol. 41 no. 2, 2000

East P,Magnan P: The effect of locomotor activityon the growth of the brook charr, Salvelinus fontinalis Mitchil!. Can. J. Zoo!. 1987,65,843-846.

Forsman L, Virtanen E: Responses of juvenile andsexually mature two-summer-old salmon (Salmosalar) to prolonged swimming. Aquaculture.1989,82, 245-255.

Foster AR, Houlihan DF, Gray C, Medale F,Fauconneau B. Kaushik SJ. Le Bail PY: The effects ofovine growth hormone on protein turnover inrainbow trout. Gen. CompoEndocrino!. 1991, 82,11-120.

Gill JA, Sumpter Jp, Donaldson EM, Dye HM, SouzaL, Berg T. Wypych 1. Langley K: Recombinantchicken and bovine growth hormones accelerategrowth in aquacultured juvenile Pacific salmonOncorhynchus kisutch . Biotechnology. 1985, 3,643-646 .

Higgs DA, Fagerlund UHF, McBride JR, Dye HM,Donaldson EM: Influence of combinations ofbovine growth hormone, 17a-methyltestosterone,and L-thyroxine on growth of yearling coho salmon (Oncorhynchus kisutch) . Can. J. Zoo!' 1977,55. 1048-1056.

Houlihan DF, Laurent P: Effects of exercise trainingon the performance, growth, and protein turnoverof rainbow trout (Salmo gairdneri) . Can. J. Fish.Aquat. Sci. 1987,44,1614-1621.

Johnston lA , Moon TW Exercise training in skeletalmuscle of brook trout (Salve linus fontinalis) . J.Exp. Bio!. 1980,87,177-194.

Jorgensen EH, Jobling M: The effects ofexercise ongrowth, food utilisation and osmoregulatory capacity of juvenile Atlantic salmon. Aquaculture .1993, 116, 233-246.

Jorgensen EH, Baardvik BM, Eliassen R, Jobling M:Food acquisition and growth of juvenile Atlanticsalmon (Salmo salar) in relation to spatial distribution offood. Aquaculture. 1996, 143, 277-289.

Kakizawa S, Kaneko T. Hasegawa S, Hirano T: Effects of feeding, fasting, background adaptation,acute stress and exhaustive exercise on theplasma somatolactin concentrations in rainbowtrout. Gen. Com, Endocrino!. 1995,98, 137-146.

Kiessling A, Higgs DA, Dosanjn BS, Eales JG : Influence of sustained exercise of two ration levels ongrowth and thyroid function of all-female Chinook salmon (Oncorhynchus tshawytscha) inseawater. Can. J. Fish. Aquat. Sci. 1994, 51,1975-1984.

Kuipers J: Salmon thrive on exercise. Fish Farmer.1982,5,9-10.


Leon KA: Effect of exerc ise on feed consumption,growth, food conversion, and stamina of brooktrout. Progressive Fish-Culturist. 1986, 48. 4346.

McKeown BA, Leatherland JF;John TM: The effectof growth hormone and prolactin on the mobilization of free fatty acids and glucose in the kokanee salmon, Oncorhynchus nerka. CompoBiochern . Physiol. 1975, 50B, 425-430.

McKinnon DL. Farrell AP: The effect of2-(thiocyanomethylthio)-benzothiazole on juvenile cohosalmon (Oncorhynchus kisutch) : sublethal toxic ity testing. Environ. Toxicol. Chern . 1992, 11,1541-1548.

McLean E, Donaldson EM. Dye HM. Souza LM:Growth acceleration of coho salmon (Oncorhynchus kisutch) following oral administration ofrecombinant bovine somatotropin. Aquaculture.1990,91.197-203.

MacLatchy DL, Kawauchi H. Eales JG: Stimulationof hepatic thyroxine 5'-deiodinase activity inrainbow trout by Pacific salmon growth hormone.CompoBiochem. Physiol. 1992, 101A, 689-691.

Moav B, McKeown BA: Thyroid hormones increasestranscription of growth hormone mRNA in rainbow trout. Hormon. Metabolic Res . 1992,24, 1014

Mommsen Tp, Vijayan MM, Moon TW Corti sol inteleosts: Dynamics, Mechanisms of action, andmetabolic regu lation . Rev. Fish BioI. 1999, 9,211-268

Nielsen ME, Boesgaard L. Sweeting RM, McKeownBA, Rosenkilde P: Plasma levels oflactate, potassium, glucose, cortisol, growth hormone, and triiodo-L-thyronine in rainbow trout (Oncorhynchus mykiss) during exercise at various levels for24 h. Can . 1. Zool. 1994,72, 1643-1647 .

Pickering AD: Progress in Comparative Endocrinology, Wiley-Liss, Inc . 1990,473-471.

Postlethwait EK, McDonald DG: Ion regulation during exercise and stress in freshwater rainbowtrout (Oncorhynchus mykiss). 1. Exp. BioI. 1995,198,295-304.

Sheridan MA: Effects of thyroxine, cortisol, growthhormone and prolactin on lipid metabolism ofcoho salmon, Oncorhynchus kisutch, duringsmoltification . Gen . Compo Endocrinol. 1986, 64.220-238.

Sheridan MA: Regulation oflipid metabolism in poikilothermic vertebrates. CompoBiochem. Physiol. 1994, 107, 495-508 .

Soivio A. Oikari A:Hematological effect s ofstress on

teleost, Esox lucius L. 1. Fish BioI. 1976,8, 39741I.

Sweeting RM, Wagner GF; McKeown BA: Changes inplasma glucose, amino acid nitrogen and growthhormone during smoltification and seawater adaptation in coho salmon, Oncorhynchus kisutch.Aquaculture. 1985,45, 185-197 .

Sweeting RM. Eales JG: The effects of fasting andfeeding on hepatic thyroxine '5-monodeiodinaseactivity in rainbow trout, Oncorhynchus mykiss .Can. 1.Zool. 1992, 70, 1516-1522 .

Totland GK, Kryvi H, Jedestel KA, Christiansen EN,Tangerds A, SUnde E: Growth and consumptionof the swimming muscle of adult Atlantic salmon, Salmo salar L., during long-term sustainedswimming. Aquaculture. 1987,66.299-313.

WalkerMG, Emerson L: Sustained swimming speedsand myotomal muscle function in trout , Salmogairdneri . 1. Fish BioI. 1978, 13, 475-48 I.

WeatherleyAH, Gill HS: Influence of bovine growthhormone on the growth dynamics ofmosaic muscle in relation to somatic growth ofrainbow trout,Salmo gairdneri Richardson. 1. Fish BioI. 1982,20. 165-172 .

Weatherley AH. Gill HS: The biology offish growth.Academic Press , London and San Diego . 1989,177-208 .

White JR, Li HW Determination ofthe energetic costofswimming from the analysis ofgrowth rate andbody composition in juvenile chinook salmon,Oncorhynchus tshawyts cha. Com . Biochem.Physiol. 1985,81,25-33.

Wootton JR: Ecology of Teleost Fishes . Chapmanand Hall, London. 1990, 404.

SammendragFysiologiske og hormonelle cendringer som felge afvedvarende svemning hos regnbueorred (Oncorhynchus mykiss) opdrcettet i akvakultur.

Regnbueerred (Oncorhynchus mykiss) blev udsat fortre niveauer af vedvarende svemning, hurtig (1,5kropsleengder S·I), moderat (0,5 kropsleengder s") ogingen (0,0 kropsleengder S·I), for at sammenligne fysiologiske og hormonelle aspekter af veekst hos erredved forskellige svemmehastigheder, Hematokrit ogplasmaniveauer af glukose, veeksthormon (GH), cortisol og triiodo-L-thyronine (T3) blev moniteret vedbegyndelsen af eksperimentet, efter 24 timer og efter2, 4, 8, 16, og 32 dages vedvarende svemning. Savelfadeindtag som morfologiske parametre blev under-

Act a vet. scand . vol. 41 no. 2, 2000


segt . Ved forseget s afslutning havde de moderatsvemmende erreder en signifikant (p<0,05) hejereveegtforegelse sammenl ignet med bade kontrolfiskene og de hurtigt svernmende fisk. De moderatsvemmende erredder konsumerede ogsa mere foderend de to andre grupper. Hematokritniveauerne stegsignifikant i bade de hurtigt og moderat svemrnendefisk ved forsegets igangsrettelse, men returnerede tilstart niveauet indenfor 8 timer. Starten af svemmeregimet forhojede plasmaniveauet af glukose, menglukose niveauet syntes herefter ikke at vrere influeret af svemningen. Ligeledes var plasrnakoncentrationerne af bade GH og T3 ikke direkte berert af

svemrneregimet, men indirekte som felge af detrendrede fedeindtag . Plasmacortisolkoncentrationensteg proportionalt med niveauet afsvemning i forbindelse med igangsrettelsen af svemmeregimet. Plasmacortisolniveauet returnerede til begyndelsesniveauetindenfor 24 timer og vedvarende svernning havdeikke nogen efterfolgende effekt pa cortisolniveauet.Resultaterne fra dette eksperiment viser, at starten paet svernmeregime rnedferer talrige rendringer iplasmakomponenterne, samt at den observeredeveegtforegelse ikke skyIdes forhejede plasmaniveauer afGH og Ty

(Received April 8, 1999; accepted February 18, 2000).

Reprints may be obtained from: M. E. Nielsen, Department of Veterinary Microbiology, Section of Fish Diseases, The Royal Veterinary and Agricultural University, Stigbejlen 4, DK-1870 Frederiksberg C, Denmark.E-mail: [email protected], tel: +45 35 282703, fax: +45 35 28 27 II.

Act a vet. scand. vol. 41 no. 2, 2000

physiological and endocrinologicalresponses

Documents